Proposal Narrative A. Project Overview Proposal.pdf · processing, spreadsheets, web navigation,...
Transcript of Proposal Narrative A. Project Overview Proposal.pdf · processing, spreadsheets, web navigation,...
Proposal Narrative
A. Project Overview
The scope of the project is to develop the course materials and laboratory setting for
students studying principles in Human-Computer Interaction (HCI) to perform an ethnographic1
design study using observation, interviews, and video tape analysis. The focus of the design
study would be students in the university's computer skills course (Applied Computer Concepts).
The computer skills course teaches a number of essential computer applications such as word
processing, spreadsheets, web navigation, web page development, and windows operating
systems. This course is taught as a combination of lecture and on-hands laboratory work. One
section of the Applied Computer Concepts course would be divided into small laboratory groups.
Each group would be assigned a student "tutor" from the HCI course. This student would
observe the group as they perform their application tasks, answering questions, and, possibly,
asking questions of the group members. Through observation and the interaction of answering
and asking questions, the HCI student would develop a user's model of the tasks being
performed, and critique how well the application's interface compliments this model. Post video
analysis of the session will help confirm and strengthen their analysis.
The HCI students will learn and practice the methodology and skills of participatory
design practice. These students will be expected to demonstrate the use of these techniques on an
interface design project for the course, as well as in their subsequent cap stone project course.
The computer skills students will have the benefit of learning in a small group setting with more
individualized attention.
1 Ethnography is a study technique in anthropology that involves the study of groups and people within the context of their everyday activities. The approach requires the researcher become a “participant-observer”, systematically recording observations and experiences [Emerson, Fretz, and Shaw]. The technique has been borrowed for the design of software applications [Huges, O’Brien, Rodden, et al.]
1
B. Goals and Objectives
The overall goal of the project is to provide the material and environment in which
computer science students have the opportunity to learn and practice participatory design
methods, and to apply these methods to human-computer interaction design problems.
C. Project Description
Design is a creative activity of making artifacts that are usable for some specific task.
Software design in particular strives for creating products that enrich the interaction between
humans and computer applications. While the software programmer/engineer is concerned with
developing reliable, robust, and maintainable software, the software designer is concerned with
creating products that fits within the user's overall activities, enhances productivity, and produces
a satisfying experience [Winograd, Kapor]. To accomplish this goal the software designer needs
to be able to apply knowledge of human goals, capabilities, and limitations with knowledge of
computer capabilities and limitations [Preece].
While it is important for a software product to provide the necessary functionality to
perform its intended use, it is also important that it presents its functionality in a manner that is
consistent with the user’s understanding. For example, the DOS operating system provided all
the necessary functionality for managing file and folders on a personal computer from a
command line. But the graphical user interfaces and the desktop metaphor of Apple’s Macintosh
and Microsoft’s Windows have transformed the personal computer operating system into a
product that could be easily used by the most non-technical users because its presentation fit
their conceptual understanding of managing files and folders.
The study of Human-Computer Interaction (HCI) has collected an array of anecdotal
evidence (e.g., Norman 88, Sachs) and significant empirical evidence (e.g., Landauer) that
2
reveals the ability for computer technology to deliver on its promises, improving our productivity
and enhancing our quality of life, rests squarely on how well the application fits our conceptual
understanding of how things work. While computer processing power and data storage has
increased dramatically, we have yet to attain a proportional increase in the productivity of work
and quality life. In part, this is because we have moved into an era where computers are being
used not simply for number crunching activities, but more to augment a variety of tasks that
humans are better suited for than machines. These are the types of tasks that cannot easily be
codified in any kind of quantitative theory, tasks that include the ability to read, understand,
negotiate, and administer [Landauer].
Left to their own devices, computer programmers take a "systems-centered point of
view", concerned about "how the software works and what parts of it do what" [Landauer, p.
217-218]. The predominant users of the number crunching era were more technically literate,
willing to put up with a high threshold of indignation (the highest level of behavioral
compromise a user is willing to make to accomplish their goals) [Saffo]. The users of the new
era are less so. They do not want to know how the inner mechanisms of the machine work; rather
they want to know how the machine will work for them. This is exactly how we need to design
such systems: the application should fit the users conception of the process, “the user-task
model”, while the inner mechanisms, “the engineer model”, should be as transparent as possible
[Gentner and Grudin]. Just as most automobile drivers (myself included) know very little of how
a car actually works, and yet find the task of driving natural, similarly we want to design
computer applications that are natural to use with little worry or concern about how they are
being accomplished.
3
In the area of HCI research, a number of approaches have evolved to meet this challenge.
These include User-Centered Design [Landauer], Human-Centered Systems [Flanagan, Huang,
Jones, and Kasif], Participatory Design [Muller and Kuhn], and Contextual Design [Beyer and
Holtzblatt]. Though they differ in their techniques, these approaches have a general common
vision of seeing "the interplay between human activity and technological systems as inextricably
linked and equally important aspects of analysis, design, and evaluation" [Flanangan, Huang,
Jones, and Kasif, p. 3]. The different techniques find ways to interject the designer in the user's
world and the user in the designer's world in order to develop a shared conceptual model of the
task and the context in which they are being done [Muller and Kuhn].
The difference between a software designer and software engineer has been compared
analogously to the difference between a building architect who designs a structure and a
contractor that builds it [Winograd]. While there are HCI degree-granting programs such as
Stanford's Center for HCI study, and larger companies such as IBM which have design and
usability labs, it is still currently the computer science programmer doing both design and
development, like an architect that both designs and constructs the building. So it is very
important that we educate computer science students in the techniques of software design that
embrace the human activity as an integral component of the analysis, design, and evaluation.
Early in the design process, the designer engages in analysis activities that provide an
insight into the user's conceptual model or mental model of the tasks for the system that is being
targeted for development [Liddle]. Mental models are cognitive artifacts that are created as we
interact with our environment that we use as a dynamic representation or simulation of our world
[Johnson-Laird]. These models "provide predictive and explanatory power for understanding the
interaction" [Norman 83]. Norman [Norman 83] distinguishes between a conceptual model,
4
which is a reasonably accurate and consistent representation of the target system, and a mental
model which is the user's cognitive representation of the target system. If the designer can devise
the conceptual model in ways that reflect a user's mental model, then the application designed
from the conceptual model will be more easily understood and fit more naturally into the user's
activities [Norman 88].
Passive observation of a user's activities is the least intrusive analysis method, and,
therefore, is the least disruptive. Yet, this approach is limited in providing insights into the user's
activities. Much of the thought process the user is going through is not verbalized. Actions or
gestures may not have apparent meanings to the observer and so are left open to a biased
interpretation. Some HCI methodologies have adapted ethnographic techniques from
anthropology that supplement passive observation with interviewing and participation [Muller
and Kuhn; Simonsen and Kensing; Rose, Shneiderman, and Plaisant; Huges, O'Brien, Rodden, et
al.; Huges, King, Rodden, Anderson]. The ethnographer develops insight by putting him or
herself into the situation they are trying to learn about. Data from a variety of sources, such as
annotations, interviews, and video, is compiled. The ethnographer interprets this data from the
context of the situation, trying to construct an understanding from the user’s point of view
[Simonsen and Kensing]. The variety of sources allows the researcher to validate the conclusions
drawn from the mass of collected data [Rose, Shneiderman, and Plaisant].
While this approach provides a data rich environment for design, it is important that it be
skillfully applied. The risk of misinterpreting observations, disrupting normal practice, and
overlooking information is high [Shneiderman]. Validated ethnographic methods have
established guidelines for performing the user study, analyzing the data, and reporting the results
[Rose, Shneiderman, and Plaisant]. Like other notable areas of computer science, these are skills
5
that must be experienced and practiced to fully learn their potential as well as their theoretical
underpinnings [Roberge, Surviano, Clarke, Fekete, and Greening]. This is one of the main goals
of this project, to provide the curriculum for an experiential learning environment for this
important aspect of HCI in designing useful and usable applications.
The "Human-Computer Interaction" course (CS 321) presents material on human aspects
of design, on technological aspects of design, and design methodologies. The laboratory
component is intended to study design in practice. Exercises include critiquing existing packages
as well as constructing interface elements. This project proposes to develop the curriculum for
the HCI students to learn and practice ethnographic analysis methods essential to creating a
user's conceptual model for design. The proposed approach would combine the HCI laboratory
experience with the laboratory experience of the "Applied Computer Concepts" course. “Applied
Computer Concepts” (CS 108) is a course taken by non-computer science majors to teach
introductory computer concepts and computer skills. The laboratory component of CS 108
teaches and practices skills of computer operating systems, word processing, spreadsheets, and
the Internet. The course assumes no prior computer experience, and is taken by a wide variety of
students.
The HCI students would "supervise and tutor", under the guidance of an instructor, one of
the laboratory skill modules to a small group of CS 108 students. The HCI student would be
available to answer questions, and have the opportunity to ask clarifying questions. Through the
interaction of observing and interviewing the CS 108 students while they work, the HCI student
can begin to develop an understanding of the naïve user’s conceptual model and determine how
well the software design fits that model. Subsequent video review will provide important
additional data that was missed during the observation session. It also aids the HCI student in
6
recognizing activities common to the individual computer users that indicate some aspect of the
conceptual model that can be generalized across the population of users they are studying. This
arrangement mirrors the way many industry software developers evaluate their products.
Lecture and exercise materials in the HCI course would provide the necessary guidelines
for conducting the user study. Beside lectures, the students would be provided examples of user
studies that would be conducted prior to the semester, which would include annotated
observation, interviews, and videotape. Both gathered data and post analysis would be presented.
This would ground the more abstract lecture material into actual experiences so that the students
will be prepared to conduct their own study.
It would be difficult or impossible to do this type of course if we were to rely on finding
40 or 50 volunteers from industry to participate as subjects in the HCI student's study. But the
Applied Computer Concepts course, CS 108, provides a natural pool of users. The students from
CS 108 also benefit. Sections of CS 108 normally have 30 students each. CS 108 students
participating in this project will have the benefit of working in a small group setting. This will
allow for more individualized attention, increased one-to-one interactions, and an atmosphere
more conducive to asking questions.
The small groups would be arranged by adding additional “special sections” of
CS 108 with strict enrollment caps of 12 students each. Each section would be
divided into three groups of 4. Four sections of CS 108 would accommodate 48
students. Each group would be taught 3 skill modules: word processing, spread
sheets, and the Internet. This would accommodate an HCI class size of up to 36,
providing each student an opportunity to teach one module (4 sections * 3 groups *
3 modules = 36).
7
Monday Tuesday Wednesday Thursday
11:30am – 12:45pm HCI Lab Section 1
HCI Lab Section 3
HCI Lab Section 1
HCI Lab Section 3
1:00pm – 2:15pm CS 108-1 CS 108-3 CS 108-1 CS 108-3
2:30pm – 3:45pm HCI Lab Section 2
HCI Lab Section 4
HCI Lab Section 2
HCI Lab Section 4
3:45pm – 5:00pm CS 108-2 CS 108-4 CS 108-2 CS 108-4
Figure 1: Example Lab Schedule
An example lab schedule is shown in Figure 1. Each HCI student would be required to
enroll in a lab section that mirrors the times of the CS 108 labs. The first part of the lab, labeled
“HCI Lab”, the HCI students would be involved in small design tasks or work on the larger
design project of the course. The second part of the lab, labeled “CS 108”, the HCI students
would be involved in the ethnographic design exercises. During any week, 12 HCI students
would be engaged in the teaching lab, the other 24 students would be involved with the other lab
experiences of the class. Each module would be taught in a 4-week segment, with three hours of
lab per week. For each HCI student, this would constitute a 12-hour commitment to this exercise,
which is only one-third of his or her lab time for the semester. During the morning and evening
hours the lab would be made available for students to work on out-of-class assignments.
The CS 108 ethnographic study is only one component of the HCI course. The course
includes material relating to other topics such as cognitive framework for design, user modeling,
and prototyping (See Figure 2). During the second half of the course, students will be given a
larger design problem to put to use the techniques they have learned. The HCI students will be
grouped into pairs and given one of two different design problems. A group with one design
problem will serve as the users for a group with the other design problem. The students will
8
perform a design analysis with their “users”, create a prototype, and perform a user testing to
determine how well their design meet the users needs. This project will in part serve as a basis
for the evaluation of the ethnographic lab exercises. The use of the participatory design methods
in the senior capstone projects will also serve as a basis for the evaluation of the HCI curriculum.
Week Lecture Topic Readings 1 Introduction - What is design? Preface & Chapter 1 of Contextual Design
Kapor, “A software design manifesto”* Gentner & Grudin, “Design models for computer human
interfaces” 2 & 3 Principles of Contextual Design -
Understanding the Customer. How to conduct an ethnography study for design.
Chapters 2 - 4 of Contextual Design Liddle, “Design of the conceptual model”*
4 & 5 User modeling - interpreting ethnographic data , mental models, cognitive framework for design.
Chapters 5 - 7 of Contextual Design Preece, Human Computer Interaction, Chapters 3 -7
6 Design specifications across multiple customers
Chapters 8 - 10 of Contextual Design Rheinfrank & Evenson, “Design languages”*
7 & 8 Designing from customer data - innovative design
Chapters 11 - 13 of Contextual Design Brown & Duguid - “Keeping it simple”* Gal, “Footholds for design”* Hammer, “Reengineering work”
9 & 10 Systems design Chapters 14 - 16 of Contextual Design Norman, “Design as practiced”* Schon & Bennett, “Reflective conversation with
materials”* 11 & 12 Prototyping - envisioning design
Design Evaluation
Chapters 17 - 19 of Contextual Design Schrage, “Cultures of prototyping”* Preece, Human-Computer Interaction, Chapter 22, 29-34
13 Design elements - interactive devices, multiple-window strategies, displaying functionality
Shneiderman, Designing the User Interface, Chapters 9, 11, & 13
14 Advanced topics in HCI - intelligent
user interfaces Designing interfaces for hearing & seeing impaired
To be announced
15 Social impact of designing user interfaces
Chapter 20 of Contextual Design Shneiderman, Designing the User Interface, Afterword
Required text books: Beyer & Holtzblatt, Contextual Design: Defining Customer-centered Systems Winograd, Bringing Design to Software (* readings in this text) Figure 2: A sample schedule of HCI course material
9
The instructors who normally teach the CS 108 courses will deliver the
lecture component of the course. In addition, the lab sections will be monitored
jointly by these instructors and the instructor of the HCI course. This will help to
insure that the CS 108 students will still receive same quality of instructions as
other course sections. While the same material will be covered because it will be
delivered in a different manner, the CS 108 students will be asked to sign an
informed consent. Further, the difference of these sections will be advertised during
the period of registration and the CS 108 students will be given an opportunity to
transfer to one of the uninvolved sections. Videotaped sessions will only be used
beyond the HCI course with additional consent given by the individuals portrayed.
The Institutional Review Board has approved this plan for projects that involve
human subjects.
Semester Sessions, 2000-2001 Summer Fall Spring Summer
Develop HCI lecture & exercise materials, set-up lab equipment, create example design study, develop CS 108 lab modules, and develop project evaluation materials.
Hire and train TA's, implement the HCI course with the new design lecture materials and lab exercises, gather data for evaluation.
Repeat the HCI course with the new design lecture materials and lab exercises, and gather data for evaluation.
Perform the analysis of the data and prepare publications and materials to disseminate results.
Figure 3: Time Line Figure 3 provides a general time line for the project. Preparation for the work will begin
during the summer session. The lab will be set-up and evaluated with the help of Matt Johnson
from Edward Jones. Mr. Johnson is a team leader for the Computer Usability Lab at Edward
Jones in St. Louis, and participates in their computer application usability studies. The School of
Engineering at SIUE has agreed to provide a room to house the lab. The room will be partitioned
10
into 4 areas. Three of these areas will house the four computers that the CS 108 users will be
performing their tasks. These will be the videotaped areas. The fourth area will contain the video
monitoring equipment that will allow for observation of the HCI and CS 108 students by
instructors in the event that any guidance is necessary but without being a constant intrusion.
This area will also house the HCI development computers that students will use to create
interfaces for their course projects.
In addition to helping develop the course materials, Dr. Mary Stephen, co-investigator
will also lend her expertise to deliver the material during the course. This approach will provide
the HCI students with the benefits of Dr. Stephen's expertise in user ethnographic studies. This
will also help to instruct current SIUE computer science faculty in the best way to teach the
material. To overcome the cross institutional difficulties of funding release time, the budget
includes consultation fees for Dr. Stephen's time to perform these duties.
With the successful completion of this project, it is our expectation that our approach and
material would be used as a model for the delivery of these important design concepts of HCI. A
tutorial video will be developed as part of the project that will be made available upon request to
other educational institutions.
The tutorial would focus on:
• The adoption of our course materials or the development of similar course materials
• The incorporation of the lab experience within an HCI course and the entire
computer science curricula.
• Structuring of the lab experience.
• How to conduct valid ethnographic studies.
• How to set-up a lab.
11
Future effort would include the development of a workshop that could be presented at the
national ACM SIGCHI conference (Special Interest Group on Computer Human-Interaction) and
the national ACM SIGCSE conference (Special Interest Group on Computer Science Education).
D. Experience of Capability of the Principal Investigators
Mary L. Stephen completed coursework in qualitative research and advanced qualitative
research as part of her Ph.D. work at St. Louis University. She has completed two substantial
qualitative research studies in recent years. One study investigated the effect on pre-service
teachers' computer anxiety when they learned software programs with first grade students. The
study has been published in the Journal of Technology and Teacher Education. She presented
results from this study at the national conference of the Association of Teacher Educators. Her
dissertation research involved the use of qualitative research methodology to study the effect of
teachers' and students' perceptions of computers in a computer-supporter classroom. This
research included a study of students' understanding of software. Some of the results have been
presented at the International Conference of the Society for Information Technology in
Education and published in the Technology and Teacher Education Annual 1998. The research
has been presented at additional conferences, including the Second International Conference on
Computer Support for Collaborative Learning. She recently participated in an exploratory study
of factors affecting students' success in beginning computer-programming courses at Southern
Illinois University at Edwardsville. In her position at Harris-Stowe State College she uses
techniques, similar to those to be used in this study, in courses and workshops designed to
prepare in-service and pre-service teachers to use technology effectively in their teaching.
Jerry Weinberg received his doctoral degree in Computer Science from Vanderbilt
University in December of 1996, holds a Bachelors of Science degree in Computer Science from
12
the University of South Carolina, and holds a Bachelors of Science degree in Nursing from
Indiana State University. His doctoral dissertation used machine learning methods from Artificial
Intelligence (AI) to extend abductive logical reasoning of medical diagnosis ("Syndromic
Abstraction: A method of exploiting domain structure to focus abductive reasoning in association
based representations "). This work also explored how the resulting process fit within a cognitive
model of diagnosis and diagnostic knowledge. The results of this work have been reported at DX
'95, an international workshop on computer diagnosis. While at Vanderbilt University, Dr.
Weinberg completed course work in cognitive science and cognitive modeling. In addition to his
research in AI and medical diagnosis, he has also done work in model-based reasoning,
qualitative modeling, and database mining. This work has been presented at various international
and national workshops and conferences. As an assistant professor at Southern Illinois
University, he worked with Dr. G. Stephen to develop the curriculum for the System Design and
Human-Computer Interaction Courses. Dr. Weinberg has taught these courses over the past 6
semesters. Dr. Weinberg is also currently teaching the Senior Capstone Project. These are all
project courses with a significant software design component.
E. Evaluation
The project will be evaluated for its effectiveness for both the HCI students (CS 321) and
the Applied Computer Concept course (CS 108). For the students in the Applied Computer
Concepts course, a quantitative study will examine the effects of the small group experience
afforded by the learning environment. This experience creates an environment that increases
exposure to a tutor, allows for more individualization of guidance, and produces an environment
that reduces apprehension to asking questions. The study will use a combination of Likert-style
surveys and comparisons of exam scores. Some of the hypotheses that will be tested are:
13
1. Applied Computer Concepts students participating will have an overall greater course
satisfaction than students in larger course sections.
2. Applied Computer Concepts students participating will by the end of the course have
reduced computer anxiety and greater confidence in their computer skills.
3. Applied Computer Concepts students participating will perform higher on
standardized computer skill exams.
The project evaluation with regards to the HCI students will answer two general
questions:
1. How well does the course material and lab experiences translate to an understanding
of design principles and practices.
2. How well does the HCI students’ understanding of design principles and practices
translate to their actual design practices.
Qualitative techniques will be used to assess these aspects of the project. The standard
criteria for judging the integrity of a quantitative research study are inconsistent with the
qualitative paradigm (Lincoln & Guba, 1985; Guba, 1981). Lincoln and Guba demonstrate that
internal validity, external validity, reliability, and objectivity all fail when applied to naturalistic
inquiries. Instead, they suggest that the following four constructs are more consistent with the
qualitative paradigm and should be used to determine the “trustworthiness” of a qualitative
study: credibility, transferability, dependability, and confirmability. Lincoln and Guba identify
strategies for dealing with the each of the criteria they list. This study will use the following
14
strategies suggested by Lincoln and Guba in order to establish the trustworthiness of the
research: (a) triangulation, (b) participant checks, and (c) audit trail.
Triangulation: “Triangulation is the act of bringing more than one source of data to bear
on a single point” (Marshall & Rossman, 1989, p. 146). This involves cross-checking
interpretations and verifying data obtained from multiple, independent approaches and sources
(Stake, 1988). The data sources for the assessment of this project will be journals kept by the
HCI students, videotapes of the sessions, and interviews with selected HCI and Applied
Computer Concept students at the end of each semester. Dr. Stephen will cross-check the data
and interpretations by pitting data obtained from one source against that obtained from another
source to confirm information and explore inconsistencies (Guba, 1981). She will look for
themes that emerge in multiple contexts.
Participant checks: Dr. Stephen will seek the participant's insights and comments on the
potential significance and role of themes and preliminary interpretations that emerge as she
gathers the data. She will do this through conversational interviews. At the conclusion of the
formal data analysis, she will discuss with selected participants her analysis of the data gathered.
Audit trail: Schofield (1989) warns that any researcher should question the truthfulness
of a study “if other researchers reading their field notes feel the evidence does not support the
way in which they have depicted the situation” (p. 93). Throughout the study, Dr. Stephen will
create an audit trail to enable other researchers to examine how she collected data, analyzed data,
and formed interpretations. She will keep a journal in which she will document these processes.
All original data, including videotapes, audiotapes, transcripts of interviews, field notes and
questionnaires will be kept in an organized and retrievable form.
15
This part of the study will focus on three particular periods of performance. First, the
student’s performance of the actual lab exercise will be evaluated. Second, student’s
performance on a course design project will be evaluated. The HCI students will be grouped into
pairs and given one of two different design problems. A group with one design problem will
serve as the users for a group with the other design problem. The students will perform a design
analysis with their “users”, create a prototype, and perform a user testing to determine how well
their design meet the users needs. Finally, the Computer Science curriculum requires each
student participate in a senior capstone project. The capstone project is a two-course sequence
where students are put into a small group of 4 to 6 and are given an large software development
project. The projects are solicited from the university and surrounding communities (examples of
projects can be found at www.cs.siue.edu/SeniorProjects). These projects have “real” users. The
students participating in this project will be followed for an extended study that will evaluate
how this earlier course affected their practices.
F. Dissemination of Results
The results of the evaluation would be of interest to the computer science education
community, the community of researchers in HCI, and to education psychologists. During the
project a World Wide Web site will be used to promote the project and provide a clearing house
of course material and results. The Web site will include the course syllabus, course materials,
results of students assignments and exercises, digital capture of lab sessions along with user
models developed by the students, and an archive of student's course projects. Details of the lab
set-up and curriculum design will also be accessible as a guide for other institutions that are
interested in doing a similar program. In addition, a tutorial videotape will be developed that
tours the lab, shows it in use, and discusses how the lab experience fits into entire HCI
16
curriculum. The copies of the videotape will be made available free of charge to educational
institutions. Research notes announcing the Web site and videotape will be sent to SIGCSE and
SIGCHI Bulletins (the major publications for these organizations). Comments from these
communities will also be solicited via an email account, which will be accessible from the Web
site. The Computer Science Department maintains it's own Web Server, so it will be easy to
keep the Web site up-to-date. The initial site will be developed by one of the supported Teaching
Assistants.
The development, implementation, and results of the HCI curriculum will be of interest
to computer science educators. The results of the study of the learning environment will be of
interest to educators of computer science educators, educators of higher learning, and educational
psychologists. Publications will be prepared for journals and conferences in these areas such as
SIGCSE, SIGCHI, and the Journal of Cognitive Psychology.
As discussed in the Project Description, additional funding would be sought to implement
the program on a wider forum. This effort would include the development of a workshop that
could be presented at the national SIGCHI conference and the national SIGCSE.
G. Equipment Request
The list of equipment request is shown in Figure 4. The proposed plan calls for a room to
be divided up into four smaller lab areas (see Figure 5 for furniture budget). Three of the areas
will be set-up with 4 computers each for the user observation studies. The videotape cameras will
be situated to capture the four user’s actions. Having three separate areas will allow for 3
sessions to be run simultaneously. This will allow for efficient monitoring of sessions by
instructors and graduate TA’s, and will allow the project to explore the potential for scaling-up to
larger HCI classes. An earlier plan called for two areas of 6 computers. But it would be difficult
17
for HCI students to observe 6 users in addition to answering the potential questions from the 2
additional users.
Videotaping is an important component for the analysis technique. To piece together a
user’s conceptual model, the designer must recognize common sets of activities that indicate a
general thought process of the user in respect to the task and computer application’s functionality
for completing the task. Direct observation and interaction is one component for gathering
information, but it is impossible to be able to make a complete set of detailed notes without
missing much of the activity occurring. A systematic videotape analysis of the sessions is the
best approach to recognizing these generalities, which are part of developing a conceptual model.
The cameras chosen are part of Panasonic’s industrial line of video equipment. They are built
with sturdier parts and casing to handle the extra wear-and-tear of heavy use. They also allow
taping with ambient light. This is important because additional bright lights would introduce a
distraction in the user’s environment. An earlier plan considered having one of these cameras be
digital to allow transfer of video to the computer. This would serve two purposes. The first is to
be able to include clips of sessions in tutorial material that would be developed for the World
Wide Web and presentation material for conference presentations. The second use of digital
video capture is for HCI students developing interfaces that may include video clips. The digital
camera technology is still quite expensive and only having one of three of the cameras seemed to
be limiting. Therefore, a good alternative is to have video capture equipment that would allow
for the digitization and editing of videotape.
The project plan includes monitoring equipment to be connected to the video cameras.
The monitoring equipment will allow instructors to view each session effectively without
introducing any undue disturbance into the user’s environment. The instructor can interject
18
advice or guidance for the HCI student during the session if it is warranted. The fourth sub-area
of the lab will house the videotaping monitors. In addition, this area will also house two
additional computers that will be set-up for interface and application development. HCI students
will be required to apply the techniques they have learned in the design study to a course design
project. These development machines will be equipped to allow the students to explore as wide a
range of interface design as possible. This may include the use of graphics, audio, and video as
elements of the interface and interaction design. Video capture equipment, as mentioned above,
would allow the digitization of video, scanning equipment would allow students to capture
graphics or forms, and audio equipment (microphones, sound cards, and speakers) will allow
students to capture audio clips. These additional two machines will have more storage and
working memory to support the applications to render and edit graphics relatively efficiently.
Equipment Item Cost
Computer Equipment
12 lab computers: 450 MHz processor, 128MB ram, audio card with speakers, Ethernet card, 17" color monitors
$2400 each
2 development computers: 450 MHz processor, 256MB ram, audio card with speakers, Ethernet card, 17" color monitors, IOMEGA Zip Drive, microphone
$2800 each
Flatbed Scanner: HP ScanJet 5100Cse $300
Adaptec VideOH!, video capture and editing system $285
Video Equipment
3 camcorders: Panasonic AG-188U $699 each
3 option batteries for camcorders: Panasonic TP-2012D $49 each
3 video tape monitors: JVC AV20920 20" color video monitors $275 each
3 interface cables: Hosa VRA-304 $10 each
3 tripods: Bogen 3165 $149 each
19
3 external camcorder microphones: Crown Sound Grabber $59 each
3 Equipment carts: Advance AV2-42E $115 each
Installation of video equipment: The AVID Group, Inc. $1,200
Supply of video tapes, $300 (to paid for by CS Dept.) $300
Figure 4: Equipment Requests
H. Equipment on Hand for the Project
The University’s Audio-Visual Services has several videotape players and monitors
available which will be used for HCI students to view and analyze their design study session.
There are 25 players available. Some of these will be reserved for the course. The number and
amount of time necessary will be determined as the need dictates. In addition to observation and
videotaping, the design study may call for user interviews. Audio-Visual Services has 15 cassette
recorders available for students to check out to support this. The University also has professional
videotape editing equipment available to faculty only. This equipment can be used to develop
tutorial material for the course and materials for workshops and conferences presentations.
I. Implementation and Equipment Maintenance
The School of Engineering has agreed to designate a room for the lab, and to purchase
the necessary furniture to support the equipment (See Figure 4). The University has received
funds to construct a new building to house all of the departments of the School of Engineering.
The Building is scheduled to be ready for occupancy in the summer semester of 2000. At that
time, funds designated for the permanent installation of equipment will be used to set-up the lab.
The Computer Science Department shares a computer support person who will oversee the
20
21
installation and maintenance of all computer equipment. In addition, the Department has put-in a
request to the State of Illinois for a full time computer support person to oversee all Department
labs and the Computer Science sub-network. Both the School of Engineering and the University
have supported this request.
In addition to the 1:1 cost sharing of the computer and video equipment, the School of
Engineering has also agreed to cost share release time during the academic year for Dr.
Weinberg. The request for release time is for one unit load for fall and spring semesters. The
School of Engineering will fund one-half of that. The School has also agreed to continue funding
the Teaching Assistant support normally allotted for the CS 108 course. This is important both
for the initial implementation of the curriculum and its continuation beyond the lifetime of the
project funding.
Funds are available from a combination of sources to support the continuation of the
curriculum after the initial funding of the project. The Computer Science Department receives an
annual equipment budget, part of that is set aside for software upgrades, hardware upgrades, and
hardware replacement. In addition, students registering for computer lab courses are charged a
nominal semester fee. This money is expressly earmarked for the maintenance and upgrade of
student labs. The combination of these sources of funding allows the Department to keep
software up-to-date, and to replace computers so that no machine is older than 3 or 4 years.
Furniture Item (To be purchased by the School of Engineering)
Cost
14 computer desks: 30 X 48 computer desk with wire management $315 each 14 chairs $88 each 8 room partitions: 60 X 66 room dividers $420 each 4 Partition wall starters $35 each Installation $600 Figure 5: Furniture Budget